Newsgroups: comp.speech
Path: cantaloupe.srv.cs.cmu.edu!rochester!galileo.cc.rochester.edu!uhura.cc.rochester.edu!rsra
From: rsra@uhura.cc.rochester.edu (Mike McCourt)
Subject: Phonetic Representation
Message-ID: <1994Oct23.004734.10046@galileo.cc.rochester.edu>
Summary: asking for experience related to the use of phonetic representation
Keywords: Speech Recognition/Genereation, Phonemes
Sender: news@galileo.cc.rochester.edu
Nntp-Posting-Host: uhura.cc.rochester.edu
Organization: University of Rochester - Rochester, New York
Date: Sun, 23 Oct 94 00:47:34 GMT
Lines: 16


I would be very glad to hear from anybody in this group that has experience
in using PHONEMIC REPRESENTATIONS for matching spoken words to their written
forms or for generating spoken words from their written forms. Is this a
valid approach or there are more effective ways for doing the same job? I
was thinking that having a phonetic representation as a buffer for passing
to the graphemic form could be of help.

I don't have access to USENET (I'm writing from the account of a friend
of mine) so please write me directly to murzaku@ux1sns.sns.it. I could send
later a summary to the comp.speech community.

Thanks,

Alex

